As the CD market has become an important source of bank funds, it has also become an important market for policymakers to understand. So far, however, model builders have not recognized the significance of assuming that new and old CDs are perfect substitutes. Therefore, they have misused the assumption, discarded relevant data, and ignored evidence inconsistent with perfect substitution. This study shows that models of the CD market should not treat new and old issues as perfect substitutes and that they should not drop observations when market rates are above the Regulation Q ceiling.
Copyright 2014 Elsevier B.V., All rights reserved.